Bad scan mirror specs?

[The_Doctor]

1200 is good, that's only £600 or so in UKP. Are they 6215's? They look longer than the 6800's. If I can get mounted 6215's for less than £700 I'll set to work on raising the money ASAP.

Nitrogen cooled large mirrors? What would those be used for? Sounds military, aiming high power at high speed.

[Pangolin]

Hi Doc,

This is taking up a lot of my time, so I will try to be brief...

Has much work been done on the idea of two drivers operating one gimballed mirror for an X/Y scan?

Yes. MIT did work in this area and used actuators from BEI Motion Systems, Kimco Magnetics division. (I did work with them many years ago too.) Actually, I did single mirror, dual-axis scanners back in around 1986, and more recently I have also consulted for another company who makes a small actuator that does this. If you take a look at US Patent number 6,720,682, you will see a device that I worked on. There are other patents that cover this same device, and if you take a very close look (by searching patents that reference this one, and the references of this patent) you will see my own scanner patents.

This particular device is smaller than a sugar cube in size, and used a 3.6mm square mirror. Sounds perfect for Lightshow, right

Since I said I will be brief, I will skip to the end of the movie and tell you the ending. Single-axis devices go faster than dual-axis devices, period. Let me tell you the ending of another movie. Rotary devices are faster than linear devices, period.

Bill, you agreed that it is desirable to get the inertias of each mirror as equal as possible, so isn't this also saying that the difference in mass is significant between the two sizes?

I don't quite know what you mean, but suffice it to say that mass does not equal inertia. Take a two-by-four and rotate it along its axis. You can rotate it very quickly in that direction -- maybe 4 or 5 times per second. Now flip it long ways and try the same thing. Now you will only be able to rotate it once every 4 or 5 seconds. Same two-by-four... same mass... different inertia.

If so, then it follows that the reduction to 3mm widths must also be significant, even with current scanners.

Nope. The difference between 1/10th the rotor inertia, and 1/15th the rotor inertia is not significant. Barely perceptible by the servo really... Also, the point of having a wider (but shorter) X axis mirror is so that the inertia IS the same. So you have two mirrors – each doing the job it was meant to do, but each having the same inertia. Everybody wins.

Either a change between the current Y mirror size and a smaller size is significant, or it is not. If it is, then I am right, If it is not, the Chinese scanner makers are right.

As a rule, the Chinese are just copying what others have done. I recently learned that one Chinese manufacturer (that we all know...) didn't even know what the pots on the scanner amp did -- I mean the names of the functions, or even how they function. They only know that if they screw around with them long enough, the ILDA test pattern should hopefully look like "this". This was an astounding discovery!!! A "manufacturer" that doesn't know what the pots do??!! Amazing. But true.

Anyway, to get back on point, no they're not right

Bill, I've thought some more on that mirror mass thing. The force conversion from electric current doesn't see a motor mass and a mirror mass, so I don't understand why you speak of power coupling efficiency on those terms. All it sees is A mass.

From a DC standpoint, maybe. But not from an AC standpoint... In another post that I will make soon, I will show you how we find resonances in scanners, and what causes the resonances. The mirror/rotor system may act like a mass-and-spring system, or even several mass-and-spring systems. The rotor is one mass, but it is also the force-producing element (together with the servo -- one of the springs). The shaft that couples the rotor to the mirror is another spring. And the mirror is a mass, but since it isn't infinitely stiff, it also has some springiness...

It can't see the difference if there is enough rigidity there.

At 20kHz, there's not much rigidity

I agree for this reason that mirrors should be stiff, but I think they should also be light.

Sure. Absolutely. But the difference between mirrors that are twice as thick is just a little bit more inertia, which still is only a small fraction of the rotor inertia. In a nutshell, it doesn't matter from a speed standpoint, especially if the limiting factor will be "ringing" in the imagery...

Smaller is the easiest way to do that, so there should be cant angle adaptations made for systems meant for faster narrower scans. This is no less relevant that using NdFeB magnets for greater flux density despite their lower heat tolerance, or using ceramic rotor formers, or any other optimisation designed to increase force to mass ratio. (Moment of inertia actually, if I remember right.)

It's KT / J, or Torque Constant to Inertia ratio. But it is Torque Constant to SYSTEM Inertia ratio. Since the mirror is such a small portion of the total SYSTEM inertia, optimization in this area does not buy you much, especially when such optimization is at the expense of something much more important – dynamic stiffness.

That CT block is actually a compromise, if it wasn't drilled with high accuracy to match the galvo body, it wouldn't be much better. If the bore was too big the points of contact would be narrow, on each side. If it was too small the contacts would be top and bottom, and although the clamping would more easily increase the area, that same clamp's leverage could easily deform the galvo body, possibly destructively if the design couldn't accept some of that pressure safely.

The outside tolerance of the scanner is controlled pretty well. That, coupled with the fact that the mount is soft aluminum, and will "conform" to the shape of the scanner when you tighten it, gives good thermal conductivity. The scanner body is made out of cold-rolled steel. It's pretty damn tough stuff... and pretty thick too. It isn't just sheet metal wrapped around the magnet!!

Actually I'd make the side walls thinner.

One thing to keep in mind. Mental exploration and debate is all well and good, but scanners and associated parts made by Cambridge, General Scanning, and other top companies were made by some damn smart people -- some of the smartest people in the world!! This is highly specialized business. These companies are rich enough to afford the best test equipment and computer software. Cambridge has done computer thermal modeling of the X-Y mount to figure out what to cut and where, and what the resulting heat dissipation is. Bottom line -- don't try to second-guess these guys!!

Best regards,

William Benner

[The_Doctor]

Since I said I will be brief, I will skip to the end of the movie and tell you the ending. Single-axis devices go faster than dual-axis devices, period. Let me tell you the ending of another movie. Rotary devices are faster than linear devices, period.

Ok, I'd guessed this much too, which is why I asked instead of trying to work on it in detail as I had on the other stuff. I didn't like the idea of one axis having to lug the mass of the other around. As for rotary, I see line as being a derivative of cycle in all things, so I'd only have considered it for cheapness if adapting speakers to open loop scanners, which I abandoned a few years ago as not being fun.

I don't quite know what you mean, but suffice it to say that mass does not equal inertia. Take a two-by-four and rotate it along its axis. You can rotate it very quickly in that direction -- maybe 4 or 5 times per second. Now flip it long ways and try the same thing. Now you will only be able to rotate it once every 4 or 5 seconds. Same two-by-four... same mass... different inertia.

I know this, but I also know that a change in mass can affect the inertia unless the shape redistributes the remaining mass around the axis to preserve that inertia. As the size reduction is a scaling operation that does not change the shape, it follows that the inertia changes. I accept that it might not be significant, I'm just making it clear that I wasn't making a gross error here.

I recently learned that one Chinese manufacturer (that we all know...) didn't even know what the pots on the scanner amp did -- I mean the names of the functions, or even how they function.

Again a chance for me to feel smug. I worked out that it would be a PID controller, but was told it actually doesn't need the Integral part, just the Proportional and Derivative, as it has to move fast and try to predict the future position for each move but not worry about the past average as it's not trying to remain steady in one setting against slow drifts in external conditions. I can't help wondering if the damping isn't still a form of Integral control though... Now, I worked out that the pots implied PID control or similar within about 36 hours of the instant I first learned that such a thing as PID control exists (this was only a few weeks ago). I am surprised that any electronics firm didn't make a similar call.

But the difference between mirrors that are twice as thick is just a little bit more inertia, which still is only a small fraction of the rotor inertia. In a nutshell, it doesn't matter from a speed standpoint, especially if the limiting factor will be "ringing" in the imagery...

Ok, but I take it that ringing is a product of resonance in the spring(s) in the system? One thing that scaling mirrors from 5 mm wide to 3 mm will do is reduce the springiness, so if that's one of the worst problems to overcome, I think it will be found useful, especially if scanner motors are made smaller, as I think they will be. I still stand by that. You could always keep the current layout and just scale the lot down, but this impacts on aperture to the point of excluding most lasers! This is why I stand by my claim, it does allow >2mm, which is generous given the mirror sizes.

But it is Torque Constant to SYSTEM Inertia ratio. Since the mirror is such a small portion of the total SYSTEM inertia, optimisation in this area does not buy you much, especially when such optimisation is at the expense of something much more important – dynamic stiffness.

I know, I made that same point about it being the whole system mass or inertia to be considered. You countered citing the mass/spring detail, yet now return to make a point I already made, right back at me as if I hadn't made it. Smaller mirrors whose thickness is either the same as before, or still in proportion as before will NOT reduce dynamic stiffness. I grant they might not make as much improvement as I'd have liked, but the idea doesn't fail as you're suggesting it does. Ringing would be slightly less, not more. The system's resonant frequency would rise, which is what is needed for increasing speed. I'll accept that given the current rotor mass the reduction won't be worth the trouble of adapting to the change, but I'm not suggesting thinner mirrors as so many are, in proportion to size, so I think you're using the argument against the Chinese choice of thinner mirrors against my idea, which isn't making their error at all.

The outside tolerance of the scanner is controlled pretty well. That, coupled with the fact that the mount is soft aluminum, and will "conform" to the shape of the scanner when you tighten it, gives good thermal conductivity. The scanner body is made out of cold-rolled steel. It's pretty damn tough stuff... and pretty thick too. It isn't just sheet metal wrapped around the magnet!!

I rest. If there is enough difference in hardness, sure, this is a good solution. I was relying on that in my diode mount too, brass being harder than the ductile Al tube I was using as a bore lining. In the end I opted for indium alloy solder in brass or copper, but I suspect that this isn't such a good idea for scanners.....

Bottom line -- don't try to second-guess these guys!!

Why not? If I hadn't dared to do that, I wouldn't learn much, let alone enough to be worth the time you've given to discuss this. If the Chinese second guessed them a bit more, they'd be making better scanners themselves. Second guessing does two things, it accelerates progress in understanding, even if it doesn't improve beyond the original; second, it might see an alternative or even an error, even if a small one. No-one's perfect. People can do what they like, even including telling another not to do something, but that doesn't mean they are right. I respect someone who's been a long way down any path looking carefully at what they see, but the next person down that path might see something new, and correctly recognise its importance even if they miss other previously recognised detail, and all who went there before should know that. As they're in a position to judge the new finding better, they should be encouraging this second guessing, not trying to deny it. If they do deny it, people will begin to wonder whose interest is really being served. I know that by taking the time to be open about this you're helping us, and that is worth more than any denial, so why encourage any?